To deal with the randomness, I had come to the conclusion that there was a seasonal arc to the GS, starting at respectable and more or less rising to strong / proficient by season's end (given careful management, which for me is in the 60 - 70 minute range). (These numbers are impressions and personal guidelines, not the result of systematic study.)
To put it another way, I base my playing time on the standard that it is very likely a given player would wind up at proficient for the playoffs if they could play 36 minutes for 2 games a week for the season.
So I like this idea there are "trends" in the GS. Anything that reinforces my own perceptions I consider to be correct.